The number and variety of clinical symptoms noted in alcohol dependent individuals typically result in a great deal of heterogeneity. Therefore, the use of diagnosis may not be optimal as a phenotypic descriptor in genetic studies of alcoholism. In this application we propose to develop fundamental endophenotypes based on neurobiological and neurobehavioral characteristics associated with alcoholism. We hypothesize that neural disinhibition is a central biologic feature in the development of alcoholism. We propose to use a novel set of neuroelectric features specifically designed to assess disinhibition in families with a high density of alcoholism. We plan to use fMRI methods to localize specific neural sources, and to conduct PET studies (Washington University and SUNY) with flumazenil to assess potential differences in binding sites (GABA) between subjects at high and low risk to develop alcoholism. We propose to assess the effects of alcohol on several measures of disinhibition using an oral alcohol challenge (UCSD), and a newly developed alcohol clamping technique (Indiana University and Howard University). In addition to the use of currently available diagnostic criteria, we will use a number of scales to assess alcohol sensitivity, tolerance, craving and dependence. Moreover, we plan to use a detailed questionnaire assessing quantity and frequency of alcohol intake as well as pattern of drinking. All of the proposed novel endophenotypes are quantitative measures which represent the essence of our IRPG. We also propose to assess the above mentioned variables in a population of African-American families. Interactive research activity dealing with standardized instruments and procedures will be carried out at seven different sites with specific shared resources described in the proposal. Finally, the eighth IRPG site (Rutgers University) will be responsible for DNA extraction and cell lines, as well as a major repository.